Online Workshop Contribution: Sosa and the Modern Position

Nearly a decade ago, Professors Curtis Bradley and Jack Goldsmith attacked the widely held “modern” position that customary international law is federal common law applicable by courts without the need for political branch authorization. This “revisionist” critique spurred extensive debate centered on four issues: (1) the historical status of CIL in the pre-Erie world of general common law, (2) the post-Erie status of CIL in the absence of political branch authorization, (3) the import of the Alien Tort Statute, in particular whether the ATS provided authorization to apply CIL as federal law, and (4) the scope of CIL norms actionable in ATS suits and the permissible sources for identifying those norms.

The Supreme Court weighed into the debate over CIL’s domestic status in its 2004 opinion in Sosa v. Alvarez-Machain. While most scholars have concluded that Sosa supports the modern position, they have reached that conclusion by conflating Sosa’s holding that congressional expectation behind the ATS supports judicial recognition of limited common law claims based on CIL (issue 3), with the separate question of whether CIL is federal common law that courts may apply without political branch authorization (issue 2). Once these issues are disentangled, Sosa cannot reasonably be read as endorsing the modern position that all CIL is automatically federal common law. The Sosa opinion is preoccupied with the limitations Erie imposed on federal common law making. Consistent with Erie, the Court found that federal courts could recognize only a small set of CIL-based common law claims and grounded the authority to do so, not in some notion that all CIL is federal common law, but in political branch authorization deriving from Congress’s expectations in enacting the ATS. The Court likewise required that the courts exercise the limited authority Congress had delegated in a manner consistent with the political branches’ intent and with those branches’ primacy in the area of foreign affairs. Finally, the Court noted that broader authority to apply CIL as federal common law would be inconsistent with Erie’s federalism concerns. The Court’s application of a traditional understanding of post-Erie federal common law to CIL is wholly inconsistent with the modern position that all CIL may be applied as federal common law without the need for political branch authorization.

Sosa’s rejection of the modern position does not mean that CIL is no longer “part of our law,” however. CIL can be adopted as, or influence, federal common law in conformance with the requirements and limitations of Erie. For example, CIL might be incorporated pursuant to a statute, such as the Foreign Sovereign Immunities Act which waives foreign states’ immunity with regard to certain claims involving property taken in violation of CIL. Similarly, CIL governing treaties might be used to interpret agreements the U.S. has ratified, with the authorization to do so proceeding from the treaty itself or from the Executive Branch which has acknowledged the binding nature of the CIL of treaties. CIL might even be borrowed to fashion domestic rules for interstate disputes pursuant to common law power emanating from the structure of the Constitution. This type of limited, interstitial common law making tied to constitutional or political branch authorization may survive Sosa’s conception of common law in a way that the modern position cannot.